Asymptotics for statistical treatment rules

Keisuke Hirano, Jack R. Porter

Research output: Contribution to journalArticle

31 Citations (Scopus)

Abstract

This paper develops asymptotic optimality theory for statistical treatment rules in smooth parametric and semiparametric models. Manski (2000, 2002, 2004) and Dehejia (2005) have argued that the problem of choosing treatments to maximize social welfare is distinct from the point estimation and hypothesis testing problems usually considered in the treatment effects literature, and advocate formal analysis of decision procedures that map empirical data into treatment choices. We develop large-sample approximations to statistical treatment assignment problems using the limits of experiments framework. We then consider some different loss functions and derive treatment assignment rules that are asymptotically optimal under average and minmax risk criteria.

Original languageEnglish (US)
Pages (from-to)1683-1701
Number of pages19
JournalEconometrica
Volume77
Issue number5
DOIs
StatePublished - Sep 2009

Fingerprint

Asymptotic optimality
Assignment problem
Semiparametric model
Parametric model
Experiment
Social welfare
Loss function
Assignment
Empirical data
Treatment effects
Approximation
Hypothesis testing

Keywords

  • Bayes rules
  • Minmax
  • Minmax regret
  • Semiparametric models
  • Statistical decision theory
  • Treatment assignment

ASJC Scopus subject areas

  • Economics and Econometrics

Cite this

Asymptotics for statistical treatment rules. / Hirano, Keisuke; Porter, Jack R.

In: Econometrica, Vol. 77, No. 5, 09.2009, p. 1683-1701.

Research output: Contribution to journalArticle

Hirano, Keisuke ; Porter, Jack R. / Asymptotics for statistical treatment rules. In: Econometrica. 2009 ; Vol. 77, No. 5. pp. 1683-1701.
@article{5128139c39d94aa4a3222378b0f6407e,
title = "Asymptotics for statistical treatment rules",
abstract = "This paper develops asymptotic optimality theory for statistical treatment rules in smooth parametric and semiparametric models. Manski (2000, 2002, 2004) and Dehejia (2005) have argued that the problem of choosing treatments to maximize social welfare is distinct from the point estimation and hypothesis testing problems usually considered in the treatment effects literature, and advocate formal analysis of decision procedures that map empirical data into treatment choices. We develop large-sample approximations to statistical treatment assignment problems using the limits of experiments framework. We then consider some different loss functions and derive treatment assignment rules that are asymptotically optimal under average and minmax risk criteria.",
keywords = "Bayes rules, Minmax, Minmax regret, Semiparametric models, Statistical decision theory, Treatment assignment",
author = "Keisuke Hirano and Porter, {Jack R.}",
year = "2009",
month = "9",
doi = "10.3982/ECTA6630",
language = "English (US)",
volume = "77",
pages = "1683--1701",
journal = "Econometrica",
issn = "0012-9682",
publisher = "Wiley-Blackwell",
number = "5",

}

TY - JOUR

T1 - Asymptotics for statistical treatment rules

AU - Hirano, Keisuke

AU - Porter, Jack R.

PY - 2009/9

Y1 - 2009/9

N2 - This paper develops asymptotic optimality theory for statistical treatment rules in smooth parametric and semiparametric models. Manski (2000, 2002, 2004) and Dehejia (2005) have argued that the problem of choosing treatments to maximize social welfare is distinct from the point estimation and hypothesis testing problems usually considered in the treatment effects literature, and advocate formal analysis of decision procedures that map empirical data into treatment choices. We develop large-sample approximations to statistical treatment assignment problems using the limits of experiments framework. We then consider some different loss functions and derive treatment assignment rules that are asymptotically optimal under average and minmax risk criteria.

AB - This paper develops asymptotic optimality theory for statistical treatment rules in smooth parametric and semiparametric models. Manski (2000, 2002, 2004) and Dehejia (2005) have argued that the problem of choosing treatments to maximize social welfare is distinct from the point estimation and hypothesis testing problems usually considered in the treatment effects literature, and advocate formal analysis of decision procedures that map empirical data into treatment choices. We develop large-sample approximations to statistical treatment assignment problems using the limits of experiments framework. We then consider some different loss functions and derive treatment assignment rules that are asymptotically optimal under average and minmax risk criteria.

KW - Bayes rules

KW - Minmax

KW - Minmax regret

KW - Semiparametric models

KW - Statistical decision theory

KW - Treatment assignment

UR - http://www.scopus.com/inward/record.url?scp=70349845924&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=70349845924&partnerID=8YFLogxK

U2 - 10.3982/ECTA6630

DO - 10.3982/ECTA6630

M3 - Article

AN - SCOPUS:70349845924

VL - 77

SP - 1683

EP - 1701

JO - Econometrica

JF - Econometrica

SN - 0012-9682

IS - 5

ER -